In [ ]:
import pandas as pd
# Load all your exported schema files (adjust file paths if needed)
clinical_demographic = pd.read_csv(r"D:\work\extra\prognosis_analysis\demographic.csv")
clinical_diagnosis = pd.read_csv(r"D:\work\extra\prognosis_analysis\diagnosis.csv")
clinical_labs = pd.read_csv(r"D:\work\extra\prognosis_analysis\labs.csv")
clinical_medication= pd.read_csv(r"D:\work\extra\prognosis_analysis\medication.csv")
clinical_vital= pd.read_csv(r"D:\work\extra\prognosis_analysis\vital_sign.csv")
clinical_urin= pd.read_csv(r"D:\work\extra\prognosis_analysis\urine.csv")
derived_heightandweight = pd.read_csv(r"D:\work\extra\prognosis_analysis\heightadnweight.csv")
derived_lab_firstday = pd.read_csv(r"D:\work\extra\prognosis_analysis\lab_firstday.csv")
derived_sofa = pd.read_csv(r"D:\work\extra\prognosis_analysis\sofa.csv")
derived_vital_firstday = pd.read_csv(r"D:\work\extra\prognosis_analysis\vital_first_day.csv")
In [ ]:
# Dictionary of DataFrames
dataframes = {
"clinical_demographic": clinical_demographic,
"clinical_diagnosis": clinical_diagnosis,
"clinical_labs": clinical_labs,
"clinical_medication": clinical_medication,
"clinical_vital": clinical_vital,
"clinical_urin": clinical_urin,
"derived_heightandweight": derived_heightandweight,
"derived_lab_firstday": derived_lab_firstday,
"derived_sofa": derived_sofa,
"derived_vital_firstday": derived_vital_firstday
}
# Display columns and data types for each DataFrame
for name, df in dataframes.items():
print(f"DataFrame: {name}")
print(df.dtypes)
print("-" * 40)
DataFrame: clinical_demographic subject_id int64 hadm_id int64 gender object ethnicity object age int64 dtype: object ---------------------------------------- DataFrame: clinical_diagnosis subject_id int64 hadm_id int64 icd9_code int64 short_title object long_title object dtype: object ---------------------------------------- DataFrame: clinical_labs subject_id int64 hadm_id int64 charttime object creatinine float64 dtype: object ---------------------------------------- DataFrame: clinical_medication subject_id int64 hadm_id int64 startdate object enddate object drug_name_generic object route object dtype: object ---------------------------------------- DataFrame: clinical_vital subject_id int64 hadm_id int64 icustay_id int64 charttime object heart_rate float64 dtype: object ---------------------------------------- DataFrame: clinical_urin subject_id int64 hadm_id int64 icustay_id int64 urine_output_24h float64 dtype: object ---------------------------------------- DataFrame: derived_heightandweight icustay_id int64 weight_first float64 weight_min float64 weight_max float64 height_first float64 height_min float64 height_max float64 dtype: object ---------------------------------------- DataFrame: derived_lab_firstday subject_id int64 hadm_id int64 icustay_id int64 ANIONGAP_min float64 ANIONGAP_max float64 ALBUMIN_min float64 ALBUMIN_max float64 BANDS_min float64 BANDS_max float64 BICARBONATE_min float64 BICARBONATE_max float64 BILIRUBIN_min float64 BILIRUBIN_max float64 CREATININE_min float64 CREATININE_max float64 CHLORIDE_min float64 CHLORIDE_max float64 GLUCOSE_min float64 GLUCOSE_max float64 HEMATOCRIT_min float64 HEMATOCRIT_max float64 HEMOGLOBIN_min float64 HEMOGLOBIN_max float64 LACTATE_min float64 LACTATE_max float64 PLATELET_min float64 PLATELET_max float64 POTASSIUM_min float64 POTASSIUM_max float64 PTT_min float64 PTT_max float64 INR_min float64 INR_max float64 PT_min float64 PT_max float64 SODIUM_min float64 SODIUM_max float64 BUN_min float64 BUN_max float64 WBC_min float64 WBC_max float64 dtype: object ---------------------------------------- DataFrame: derived_sofa subject_id int64 hadm_id int64 icustay_id int64 SOFA int64 respiration float64 coagulation float64 liver float64 cardiovascular float64 cns float64 renal float64 dtype: object ---------------------------------------- DataFrame: derived_vital_firstday subject_id int64 hadm_id int64 icustay_id int64 HeartRate_Min float64 HeartRate_Max float64 HeartRate_Mean float64 SysBP_Min float64 SysBP_Max float64 SysBP_Mean float64 DiasBP_Min float64 DiasBP_Max float64 DiasBP_Mean float64 MeanBP_Min float64 MeanBP_Max float64 MeanBP_Mean float64 RespRate_Min float64 RespRate_Max float64 RespRate_Mean float64 TempC_Min float64 TempC_Max float64 TempC_Mean float64 SpO2_Min float64 SpO2_Max float64 SpO2_Mean float64 Glucose_Min float64 Glucose_Max float64 Glucose_Mean float64 dtype: object ----------------------------------------
In [ ]:
import pandas as pd
# STEP 1: Start with clinical_demographic as the base
master_df = clinical_demographic.copy()
# STEP 2: Merge other dataframes using subject_id and hadm_id
master_df = master_df.merge(clinical_urin, on=['subject_id', 'hadm_id'], how='left')
master_df = master_df.merge(clinical_labs, on=['subject_id', 'hadm_id'], how='left')
master_df = master_df.merge(clinical_vital, on=['subject_id', 'hadm_id'], how='left')
# STEP 3: Merge derived tables using subject_id and hadm_id
master_df = master_df.merge(derived_lab_firstday, on=['subject_id', 'hadm_id'], how='left')
# Clean derived_sofa to avoid duplicate 'icustay_id'
derived_sofa_clean = derived_sofa.drop(columns=['icustay_id'], errors='ignore')
master_df = master_df.merge(derived_sofa_clean, on=['subject_id', 'hadm_id'], how='left')
# STEP 4: Merge vitals (first day summary)
derived_vital_firstday_clean = derived_vital_firstday.drop(columns=['icustay_id'], errors='ignore')
master_df = master_df.merge(derived_vital_firstday_clean, on=['subject_id', 'hadm_id'], how='left')
# STEP 5: Now merge derived height/weight using icustay_id
# Ensure icustay_id exists in master_df before merging
if 'icustay_id' in master_df.columns:
master_df = master_df.merge(derived_heightandweight, on='icustay_id', how='left')
else:
print("⚠️ 'icustay_id' not found in master_df; skipping height/weight merge.")
# Final sanity check
print(f"✅ Final master_df shape: {master_df.shape}")
print("🧾 Final columns:")
print(master_df.columns.tolist())
✅ Final master_df shape: (1114581, 88) 🧾 Final columns: ['subject_id', 'hadm_id', 'gender', 'ethnicity', 'age', 'icustay_id_x', 'urine_output_24h', 'charttime_x', 'creatinine', 'icustay_id_y', 'charttime_y', 'heart_rate', 'icustay_id', 'ANIONGAP_min', 'ANIONGAP_max', 'ALBUMIN_min', 'ALBUMIN_max', 'BANDS_min', 'BANDS_max', 'BICARBONATE_min', 'BICARBONATE_max', 'BILIRUBIN_min', 'BILIRUBIN_max', 'CREATININE_min', 'CREATININE_max', 'CHLORIDE_min', 'CHLORIDE_max', 'GLUCOSE_min', 'GLUCOSE_max', 'HEMATOCRIT_min', 'HEMATOCRIT_max', 'HEMOGLOBIN_min', 'HEMOGLOBIN_max', 'LACTATE_min', 'LACTATE_max', 'PLATELET_min', 'PLATELET_max', 'POTASSIUM_min', 'POTASSIUM_max', 'PTT_min', 'PTT_max', 'INR_min', 'INR_max', 'PT_min', 'PT_max', 'SODIUM_min', 'SODIUM_max', 'BUN_min', 'BUN_max', 'WBC_min', 'WBC_max', 'SOFA', 'respiration', 'coagulation', 'liver', 'cardiovascular', 'cns', 'renal', 'HeartRate_Min', 'HeartRate_Max', 'HeartRate_Mean', 'SysBP_Min', 'SysBP_Max', 'SysBP_Mean', 'DiasBP_Min', 'DiasBP_Max', 'DiasBP_Mean', 'MeanBP_Min', 'MeanBP_Max', 'MeanBP_Mean', 'RespRate_Min', 'RespRate_Max', 'RespRate_Mean', 'TempC_Min', 'TempC_Max', 'TempC_Mean', 'SpO2_Min', 'SpO2_Max', 'SpO2_Mean', 'Glucose_Min', 'Glucose_Max', 'Glucose_Mean', 'weight_first', 'weight_min', 'weight_max', 'height_first', 'height_min', 'height_max']
In [ ]:
master_df.describe(include='all')
Out[ ]:
| subject_id | hadm_id | gender | ethnicity | age | icustay_id_x | urine_output_24h | charttime_x | creatinine | icustay_id_y | ... | SpO2_Mean | Glucose_Min | Glucose_Max | Glucose_Mean | weight_first | weight_min | weight_max | height_first | height_min | height_max | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| count | 1.114581e+06 | 1.114581e+06 | 1114581 | 1114581 | 1.114581e+06 | 718332.000000 | 718332.000000 | 1040416 | 1.040416e+06 | 1.011682e+06 | ... | 870212.000000 | 870763.000000 | 870763.000000 | 870763.000000 | 938978.000000 | 938978.000000 | 9.389780e+05 | 370042.000000 | 370149.000000 | 370149.000000 |
| unique | NaN | NaN | 2 | 41 | NaN | NaN | NaN | 97499 | NaN | NaN | ... | NaN | NaN | NaN | NaN | NaN | NaN | NaN | NaN | NaN | NaN |
| top | NaN | NaN | M | WHITE | NaN | NaN | NaN | 2173-04-09T11:55:00 | NaN | NaN | ... | NaN | NaN | NaN | NaN | NaN | NaN | NaN | NaN | NaN | NaN |
| freq | NaN | NaN | 731330 | 740421 | NaN | NaN | NaN | 5100 | NaN | NaN | ... | NaN | NaN | NaN | NaN | NaN | NaN | NaN | NaN | NaN | NaN |
| mean | 4.969088e+04 | 1.488176e+05 | NaN | NaN | 5.723565e+01 | 249059.968070 | 3786.518068 | NaN | 1.779993e+00 | 2.476294e+05 | ... | 97.456483 | 105.850153 | 291.028124 | 155.596661 | 76.961322 | 74.779984 | 3.388204e+02 | 151.138792 | 151.079339 | 151.568107 |
| std | 3.126975e+04 | 2.957085e+04 | NaN | NaN | 4.136289e+01 | 28344.672879 | 7165.135564 | NaN | 1.650737e+00 | 2.941916e+04 | ... | 2.597804 | 39.114159 | 9093.028723 | 1299.485333 | 30.054365 | 28.923110 | 4.702277e+04 | 46.001403 | 46.128332 | 45.026286 |
| min | 2.000000e+00 | 1.000010e+05 | NaN | NaN | 0.000000e+00 | 200001.000000 | 0.000000 | NaN | 0.000000e+00 | 2.000030e+05 | ... | 11.166667 | 0.830000 | 9.000000 | 7.000000 | 0.360000 | 0.070000 | 3.700000e-01 | 14.000000 | 5.000000 | 17.000000 |
| 25% | 2.204000e+04 | 1.231780e+05 | NaN | NaN | 3.900000e+01 | 228140.000000 | 1293.000000 | NaN | 8.000000e-01 | 2.230560e+05 | ... | 96.600000 | 79.000000 | 129.000000 | 109.750000 | 62.000000 | 60.330000 | 6.570000e+01 | 154.940000 | 154.940000 | 154.940000 |
| 50% | 4.847900e+04 | 1.439970e+05 | NaN | NaN | 5.800000e+01 | 248745.000000 | 2220.000000 | NaN | 1.200000e+00 | 2.467160e+05 | ... | 97.880000 | 102.000000 | 175.000000 | 135.888889 | 77.600000 | 75.800000 | 8.150000e+01 | 167.640000 | 167.640000 | 167.640000 |
| 75% | 7.596000e+04 | 1.732830e+05 | NaN | NaN | 7.000000e+01 | 269593.000000 | 4180.000000 | NaN | 2.000000e+00 | 2.707400e+05 | ... | 99.136364 | 126.000000 | 214.000000 | 170.400000 | 90.000000 | 88.100000 | 9.900000e+01 | 177.800000 | 177.800000 | 177.800000 |
| max | 9.999900e+04 | 1.999990e+05 | NaN | NaN | 3.110000e+02 | 299998.000000 | 561190.000000 | NaN | 4.660000e+01 | 2.999930e+05 | ... | 100.000000 | 592.000000 | 999999.000000 | 142966.857143 | 710.400000 | 575.000000 | 8.654765e+06 | 210.820000 | 210.820000 | 210.820000 |
11 rows × 88 columns
In [ ]:
# Load patients table
patients = pd.read_csv(r"D:\work\extra\prognosis_analysis\pateint.csv", parse_dates=["dob", "dod"])
# Merge date of death into your master
master_df = master_df.merge(patients[['subject_id', 'dod']], on='subject_id', how='left')
In [ ]:
master_df.columns
Out[ ]:
Index(['subject_id', 'hadm_id', 'gender', 'ethnicity', 'age', 'icustay_id_x',
'urine_output_24h', 'charttime_x', 'creatinine', 'icustay_id_y',
'charttime_y', 'heart_rate', 'icustay_id', 'ANIONGAP_min',
'ANIONGAP_max', 'ALBUMIN_min', 'ALBUMIN_max', 'BANDS_min', 'BANDS_max',
'BICARBONATE_min', 'BICARBONATE_max', 'BILIRUBIN_min', 'BILIRUBIN_max',
'CREATININE_min', 'CREATININE_max', 'CHLORIDE_min', 'CHLORIDE_max',
'GLUCOSE_min', 'GLUCOSE_max', 'HEMATOCRIT_min', 'HEMATOCRIT_max',
'HEMOGLOBIN_min', 'HEMOGLOBIN_max', 'LACTATE_min', 'LACTATE_max',
'PLATELET_min', 'PLATELET_max', 'POTASSIUM_min', 'POTASSIUM_max',
'PTT_min', 'PTT_max', 'INR_min', 'INR_max', 'PT_min', 'PT_max',
'SODIUM_min', 'SODIUM_max', 'BUN_min', 'BUN_max', 'WBC_min', 'WBC_max',
'SOFA', 'respiration', 'coagulation', 'liver', 'cardiovascular', 'cns',
'renal', 'HeartRate_Min', 'HeartRate_Max', 'HeartRate_Mean',
'SysBP_Min', 'SysBP_Max', 'SysBP_Mean', 'DiasBP_Min', 'DiasBP_Max',
'DiasBP_Mean', 'MeanBP_Min', 'MeanBP_Max', 'MeanBP_Mean',
'RespRate_Min', 'RespRate_Max', 'RespRate_Mean', 'TempC_Min',
'TempC_Max', 'TempC_Mean', 'SpO2_Min', 'SpO2_Max', 'SpO2_Mean',
'Glucose_Min', 'Glucose_Max', 'Glucose_Mean', 'weight_first',
'weight_min', 'weight_max', 'height_first', 'height_min', 'height_max',
'dod'],
dtype='object')
In [ ]:
master_df['bmi'] = master_df['weight_first'] / ((master_df['height_first'] / 100) ** 2)
In [ ]:
import numpy as np
from datetime import datetime
# Convert datetimes
master_df['admittime'] = pd.to_datetime(master_df['charttime_x'])
master_df['dod'] = pd.to_datetime(master_df['dod'])
# Calculate days to death
master_df['days_to_death'] = (master_df['dod'] - master_df['admittime']).dt.days
# Define outcome class
def classify_survival(days):
if pd.isna(days):
return 3 # survivors or unknown
elif days < 30:
return 0
elif days < 365:
return 1
elif days < 5*365:
return 2
else:
return 3
master_df['outcome_class'] = master_df['days_to_death'].apply(classify_survival)
In [ ]:
master_df['admittime'] = pd.to_datetime(master_df['charttime_x'])
master_df['dod'] = pd.to_datetime(master_df['dod'])
In [ ]:
def classify_survival(days):
if pd.isna(days): return 3
elif days < 30: return 0
elif days < 365: return 1
elif days < 5*365: return 2
else: return 3
master_df['outcome_class'] = master_df['days_to_death'].apply(classify_survival)
In [ ]:
# Drop missing survival classes or key fields
analysis_df = master_df.dropna(subset=['outcome_class'])
In [ ]:
analysis_df
Out[ ]:
| subject_id | hadm_id | gender | ethnicity | age | icustay_id_x | urine_output_24h | charttime_x | creatinine | icustay_id_y | ... | weight_min | weight_max | height_first | height_min | height_max | dod | bmi | admittime | days_to_death | outcome_class | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 0 | 88 | 123010 | M | BLACK/AFRICAN AMERICAN | 29 | NaN | NaN | 2111-08-29T04:04:00 | 0.9 | 297289.0 | ... | 95.2 | 95.2 | NaN | NaN | NaN | NaT | NaN | 2111-08-29 04:04:00 | NaN | 3 |
| 1 | 88 | 123010 | M | BLACK/AFRICAN AMERICAN | 29 | NaN | NaN | 2111-08-30T01:23:00 | 0.7 | 297289.0 | ... | 95.2 | 95.2 | NaN | NaN | NaN | NaT | NaN | 2111-08-30 01:23:00 | NaN | 3 |
| 2 | 855 | 173950 | F | UNKNOWN/NOT SPECIFIED | 33 | NaN | NaN | 2138-06-27T03:24:00 | 0.5 | NaN | ... | NaN | NaN | NaN | NaN | NaN | NaT | NaN | 2138-06-27 03:24:00 | NaN | 3 |
| 3 | 748 | 171044 | M | UNKNOWN/NOT SPECIFIED | 20 | NaN | NaN | 2101-09-19T02:15:00 | 0.8 | 222896.0 | ... | NaN | NaN | NaN | NaN | NaN | NaT | NaN | 2101-09-19 02:15:00 | NaN | 3 |
| 4 | 748 | 171044 | M | UNKNOWN/NOT SPECIFIED | 20 | NaN | NaN | 2101-09-19T02:15:00 | 0.8 | 222896.0 | ... | NaN | NaN | NaN | NaN | NaN | NaT | NaN | 2101-09-19 02:15:00 | NaN | 3 |
| ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... |
| 2280124 | 17918 | 101826 | M | WHITE | 300 | NaN | NaN | 2127-12-18T07:02:00 | 1.0 | 212791.0 | ... | NaN | NaN | NaN | NaN | NaN | 2127-12-25 | NaN | 2127-12-18 07:02:00 | 6.0 | 0 |
| 2280125 | 17918 | 101826 | M | WHITE | 300 | NaN | NaN | 2127-12-18T07:02:00 | 1.0 | 212791.0 | ... | NaN | NaN | NaN | NaN | NaN | 2127-12-25 | NaN | 2127-12-18 07:02:00 | 6.0 | 0 |
| 2280126 | 17918 | 101826 | M | WHITE | 300 | NaN | NaN | 2127-12-18T07:02:00 | 1.0 | 212791.0 | ... | NaN | NaN | NaN | NaN | NaN | 2127-12-25 | NaN | 2127-12-18 07:02:00 | 6.0 | 0 |
| 2280127 | 17918 | 101826 | M | WHITE | 300 | NaN | NaN | 2127-12-18T07:02:00 | 1.0 | 212791.0 | ... | NaN | NaN | NaN | NaN | NaN | 2127-12-25 | NaN | 2127-12-18 07:02:00 | 6.0 | 0 |
| 2280128 | 17918 | 101826 | M | WHITE | 300 | NaN | NaN | 2127-12-18T07:02:00 | 1.0 | 212791.0 | ... | NaN | NaN | NaN | NaN | NaN | 2127-12-25 | NaN | 2127-12-18 07:02:00 | 6.0 | 0 |
2280129 rows × 93 columns
In [ ]:
# Initialize existing lists
continuous_vars = []
categorical_vars = []
# Check dtypes and value counts to classify variables
for col in analysis_df.columns:
if col in continuous_vars or col in categorical_vars:
continue # Already classified
if pd.api.types.is_numeric_dtype(analysis_df[col]):
unique_vals = analysis_df[col].nunique()
if unique_vals < 10: # Likely a categorical numeric (e.g. 0/1, codes)
categorical_vars.append(col)
else:
continuous_vars.append(col)
else:
categorical_vars.append(col)
print("Continuous Variables:", continuous_vars)
print("Categorical Variables:", categorical_vars)
Continuous Variables: ['subject_id', 'hadm_id', 'age', 'icustay_id_x', 'urine_output_24h', 'creatinine', 'icustay_id_y', 'heart_rate', 'icustay_id', 'ANIONGAP_min', 'ANIONGAP_max', 'ALBUMIN_min', 'ALBUMIN_max', 'BANDS_min', 'BANDS_max', 'BICARBONATE_min', 'BICARBONATE_max', 'BILIRUBIN_min', 'BILIRUBIN_max', 'CREATININE_min', 'CREATININE_max', 'CHLORIDE_min', 'CHLORIDE_max', 'GLUCOSE_min', 'GLUCOSE_max', 'HEMATOCRIT_min', 'HEMATOCRIT_max', 'HEMOGLOBIN_min', 'HEMOGLOBIN_max', 'LACTATE_min', 'LACTATE_max', 'PLATELET_min', 'PLATELET_max', 'POTASSIUM_min', 'POTASSIUM_max', 'PTT_min', 'PTT_max', 'INR_min', 'INR_max', 'PT_min', 'PT_max', 'SODIUM_min', 'SODIUM_max', 'BUN_min', 'BUN_max', 'WBC_min', 'WBC_max', 'SOFA', 'HeartRate_Min', 'HeartRate_Max', 'HeartRate_Mean', 'SysBP_Min', 'SysBP_Max', 'SysBP_Mean', 'DiasBP_Min', 'DiasBP_Max', 'DiasBP_Mean', 'MeanBP_Min', 'MeanBP_Max', 'MeanBP_Mean', 'RespRate_Min', 'RespRate_Max', 'RespRate_Mean', 'TempC_Min', 'TempC_Max', 'TempC_Mean', 'SpO2_Min', 'SpO2_Max', 'SpO2_Mean', 'Glucose_Min', 'Glucose_Max', 'Glucose_Mean', 'weight_first', 'weight_min', 'weight_max', 'height_first', 'height_min', 'height_max', 'bmi', 'days_to_death'] Categorical Variables: ['gender', 'ethnicity', 'charttime_x', 'charttime_y', 'respiration', 'coagulation', 'liver', 'cardiovascular', 'cns', 'renal', 'dod', 'admittime', 'outcome_class']
In [ ]:
from scipy.stats import shapiro,kruskal,f_oneway
from tabulate import tabulate
print("\n📊 CONTINUOUS VARIABLES ANALYSIS\n")
for var in continuous_vars:
print(f"\n--- {var} ---")
# Drop NaNs
sub_df = analysis_df.dropna(subset=[var])
# Check normality
p_norm = shapiro(sub_df[var])[1]
is_normal = p_norm > 0.05
# Group data by survival class
samples = [group[var].dropna() for _, group in sub_df.groupby('outcome_class')]
# Summary stats
stats_by_group = []
for group_num, values in enumerate(samples):
mean = values.mean()
std = values.std()
median = values.median()
q1 = values.quantile(0.25)
q3 = values.quantile(0.75)
stats_by_group.append([group_num,
f"{mean:.2f} ± {std:.2f}" if is_normal else f"{median:.2f} [{q1:.2f}–{q3:.2f}]"])
# Statistical test
if is_normal:
_, p = f_oneway(*samples)
test_name = "ANOVA"
else:
_, p = kruskal(*samples)
test_name = "Kruskal-Wallis"
print(tabulate(stats_by_group, headers=["Class", "Summary"], tablefmt="github"))
print(f"{test_name} p = {p:.4f} {'*' if p < 0.05 else ''}")
📊 CONTINUOUS VARIABLES ANALYSIS --- subject_id ---
c:\Users\home\anaconda3\envs\project1\Lib\site-packages\scipy\stats\_axis_nan_policy.py:531: UserWarning: scipy.stats.shapiro: For N > 5000, computed p-value may not be accurate. Current N is 2280129. res = hypotest_fun_out(*samples, **kwds)
| Class | Summary | |---------|------------------------------| | 0 | 58505.00 [21122.00–81719.00] | | 1 | 76327.00 [55921.00–98220.00] | | 2 | 25699.00 [17891.00–31779.00] | | 3 | 28824.00 [19533.00–55094.00] | Kruskal-Wallis p = 0.0000 * --- hadm_id ---
c:\Users\home\anaconda3\envs\project1\Lib\site-packages\scipy\stats\_axis_nan_policy.py:531: UserWarning: scipy.stats.shapiro: For N > 5000, computed p-value may not be accurate. Current N is 2280129. res = hypotest_fun_out(*samples, **kwds)
| Class | Summary | |---------|---------------------------------| | 0 | 143320.00 [117847.00–175729.00] | | 1 | 173136.00 [129611.00–173136.00] | | 2 | 164768.00 [139238.00–186706.00] | | 3 | 146210.00 [123178.00–183190.00] | Kruskal-Wallis p = 0.0000 * --- age ---
c:\Users\home\anaconda3\envs\project1\Lib\site-packages\scipy\stats\_axis_nan_policy.py:531: UserWarning: scipy.stats.shapiro: For N > 5000, computed p-value may not be accurate. Current N is 2280129. res = hypotest_fun_out(*samples, **kwds)
| Class | Summary | |---------|---------------------| | 0 | 64.00 [55.00–72.00] | | 1 | 70.00 [43.00–70.00] | | 2 | 51.00 [39.00–67.00] | | 3 | 51.00 [34.00–61.00] | Kruskal-Wallis p = 0.0000 * --- icustay_id_x ---
c:\Users\home\anaconda3\envs\project1\Lib\site-packages\scipy\stats\_axis_nan_policy.py:531: UserWarning: scipy.stats.shapiro: For N > 5000, computed p-value may not be accurate. Current N is 1496791. res = hypotest_fun_out(*samples, **kwds)
| Class | Summary | |---------|---------------------------------| | 0 | 254019.00 [230949.00–276413.00] | | 1 | 243420.00 [228831.00–265955.00] | | 2 | 244441.00 [215438.00–268865.00] | | 3 | 247086.00 [219648.00–271393.00] | Kruskal-Wallis p = 0.0000 * --- urine_output_24h ---
c:\Users\home\anaconda3\envs\project1\Lib\site-packages\scipy\stats\_axis_nan_policy.py:531: UserWarning: scipy.stats.shapiro: For N > 5000, computed p-value may not be accurate. Current N is 1496791. res = hypotest_fun_out(*samples, **kwds)
| Class | Summary | |---------|---------------------------| | 0 | 1587.00 [750.00–3065.00] | | 1 | 2925.00 [1255.00–4280.00] | | 2 | 2140.00 [1100.00–3145.00] | | 3 | 2700.00 [1750.00–4400.00] | Kruskal-Wallis p = 0.0000 * --- creatinine ---
c:\Users\home\anaconda3\envs\project1\Lib\site-packages\scipy\stats\_axis_nan_policy.py:531: UserWarning: scipy.stats.shapiro: For N > 5000, computed p-value may not be accurate. Current N is 2197156. res = hypotest_fun_out(*samples, **kwds)
| Class | Summary | |---------|------------------| | 0 | 1.60 [0.90–2.60] | | 1 | 1.80 [1.20–3.10] | | 2 | 1.10 [0.80–2.00] | | 3 | 0.90 [0.60–1.50] | Kruskal-Wallis p = 0.0000 * --- icustay_id_y ---
c:\Users\home\anaconda3\envs\project1\Lib\site-packages\scipy\stats\_axis_nan_policy.py:531: UserWarning: scipy.stats.shapiro: For N > 5000, computed p-value may not be accurate. Current N is 2060173. res = hypotest_fun_out(*samples, **kwds)
| Class | Summary | |---------|---------------------------------| | 0 | 255459.00 [233883.00–273412.00] | | 1 | 235538.00 [234993.00–271398.00] | | 2 | 235344.00 [224986.00–262256.00] | | 3 | 240659.00 [213694.00–265939.00] | Kruskal-Wallis p = 0.0000 * --- heart_rate ---
c:\Users\home\anaconda3\envs\project1\Lib\site-packages\scipy\stats\_axis_nan_policy.py:531: UserWarning: scipy.stats.shapiro: For N > 5000, computed p-value may not be accurate. Current N is 2060173. res = hypotest_fun_out(*samples, **kwds)
| Class | Summary | |---------|------------------------| | 0 | 127.00 [121.00–133.00] | | 1 | 123.00 [121.00–129.00] | | 2 | 56.00 [51.00–127.00] | | 3 | 123.00 [55.00–132.00] | Kruskal-Wallis p = 0.0000 * --- icustay_id ---
c:\Users\home\anaconda3\envs\project1\Lib\site-packages\scipy\stats\_axis_nan_policy.py:531: UserWarning: scipy.stats.shapiro: For N > 5000, computed p-value may not be accurate. Current N is 1983728. res = hypotest_fun_out(*samples, **kwds)
| Class | Summary | |---------|---------------------------------| | 0 | 254019.00 [228140.00–275736.00] | | 1 | 243420.00 [234993.00–265955.00] | | 2 | 257032.00 [226377.00–271527.00] | | 3 | 244135.00 [217070.00–267478.00] | Kruskal-Wallis p = 0.0000 * --- ANIONGAP_min ---
c:\Users\home\anaconda3\envs\project1\Lib\site-packages\scipy\stats\_axis_nan_policy.py:531: UserWarning: scipy.stats.shapiro: For N > 5000, computed p-value may not be accurate. Current N is 1933550. res = hypotest_fun_out(*samples, **kwds)
| Class | Summary | |---------|---------------------| | 0 | 15.00 [12.00–18.00] | | 1 | 13.00 [12.00–16.00] | | 2 | 13.00 [12.00–15.00] | | 3 | 13.00 [11.00–15.00] | Kruskal-Wallis p = 0.0000 * --- ANIONGAP_max ---
c:\Users\home\anaconda3\envs\project1\Lib\site-packages\scipy\stats\_axis_nan_policy.py:531: UserWarning: scipy.stats.shapiro: For N > 5000, computed p-value may not be accurate. Current N is 1933550. res = hypotest_fun_out(*samples, **kwds)
| Class | Summary | |---------|---------------------| | 0 | 19.00 [15.00–23.00] | | 1 | 16.00 [15.00–19.00] | | 2 | 17.00 [15.00–19.00] | | 3 | 16.00 [14.00–20.00] | Kruskal-Wallis p = 0.0000 * --- ALBUMIN_min ---
c:\Users\home\anaconda3\envs\project1\Lib\site-packages\scipy\stats\_axis_nan_policy.py:531: UserWarning: scipy.stats.shapiro: For N > 5000, computed p-value may not be accurate. Current N is 1318511. res = hypotest_fun_out(*samples, **kwds)
| Class | Summary | |---------|------------------| | 0 | 2.50 [2.00–2.80] | | 1 | 2.50 [2.10–2.70] | | 2 | 2.60 [2.20–2.90] | | 3 | 2.30 [1.90–2.70] | Kruskal-Wallis p = 0.0000 * --- ALBUMIN_max ---
c:\Users\home\anaconda3\envs\project1\Lib\site-packages\scipy\stats\_axis_nan_policy.py:531: UserWarning: scipy.stats.shapiro: For N > 5000, computed p-value may not be accurate. Current N is 1318511. res = hypotest_fun_out(*samples, **kwds)
| Class | Summary | |---------|------------------| | 0 | 2.50 [2.20–3.00] | | 1 | 2.50 [2.20–2.70] | | 2 | 2.70 [2.30–3.00] | | 3 | 2.50 [2.20–3.00] | Kruskal-Wallis p = 0.0000 * --- BANDS_min ---
c:\Users\home\anaconda3\envs\project1\Lib\site-packages\scipy\stats\_axis_nan_policy.py:531: UserWarning: scipy.stats.shapiro: For N > 5000, computed p-value may not be accurate. Current N is 327855. res = hypotest_fun_out(*samples, **kwds)
| Class | Summary | |---------|-------------------| | 0 | 5.00 [2.00–8.00] | | 1 | 5.00 [2.00–18.00] | | 2 | 6.00 [1.00–13.00] | | 3 | 6.00 [2.00–18.00] | Kruskal-Wallis p = 0.0000 * --- BANDS_max --- | Class | Summary | |---------|-------------------| | 0 | 5.00 [2.00–11.00] | | 1 | 5.00 [2.00–20.00] | | 2 | 7.00 [1.00–13.00] | | 3 | 8.00 [2.00–22.00] | Kruskal-Wallis p = 0.0000 * --- BICARBONATE_min ---
c:\Users\home\anaconda3\envs\project1\Lib\site-packages\scipy\stats\_axis_nan_policy.py:531: UserWarning: scipy.stats.shapiro: For N > 5000, computed p-value may not be accurate. Current N is 327855. res = hypotest_fun_out(*samples, **kwds) c:\Users\home\anaconda3\envs\project1\Lib\site-packages\scipy\stats\_axis_nan_policy.py:531: UserWarning: scipy.stats.shapiro: For N > 5000, computed p-value may not be accurate. Current N is 1935734. res = hypotest_fun_out(*samples, **kwds)
| Class | Summary | |---------|---------------------| | 0 | 17.00 [14.00–20.00] | | 1 | 19.00 [16.00–21.00] | | 2 | 19.00 [15.00–20.00] | | 3 | 20.00 [18.00–21.00] | Kruskal-Wallis p = 0.0000 * --- BICARBONATE_max ---
c:\Users\home\anaconda3\envs\project1\Lib\site-packages\scipy\stats\_axis_nan_policy.py:531: UserWarning: scipy.stats.shapiro: For N > 5000, computed p-value may not be accurate. Current N is 1935734. res = hypotest_fun_out(*samples, **kwds)
| Class | Summary | |---------|---------------------| | 0 | 21.00 [19.00–23.00] | | 1 | 22.00 [20.00–22.00] | | 2 | 21.00 [19.00–24.00] | | 3 | 22.00 [21.00–25.00] | Kruskal-Wallis p = 0.0000 * --- BILIRUBIN_min ---
c:\Users\home\anaconda3\envs\project1\Lib\site-packages\scipy\stats\_axis_nan_policy.py:531: UserWarning: scipy.stats.shapiro: For N > 5000, computed p-value may not be accurate. Current N is 1599847. res = hypotest_fun_out(*samples, **kwds)
| Class | Summary | |---------|-------------------| | 0 | 1.10 [0.70–5.10] | | 1 | 2.20 [0.60–5.40] | | 2 | 3.30 [1.30–16.00] | | 3 | 1.70 [0.50–17.40] | Kruskal-Wallis p = 0.0000 * --- BILIRUBIN_max ---
c:\Users\home\anaconda3\envs\project1\Lib\site-packages\scipy\stats\_axis_nan_policy.py:531: UserWarning: scipy.stats.shapiro: For N > 5000, computed p-value may not be accurate. Current N is 1599847. res = hypotest_fun_out(*samples, **kwds)
| Class | Summary | |---------|-------------------| | 0 | 1.60 [0.70–6.30] | | 1 | 2.20 [0.70–5.40] | | 2 | 5.30 [1.50–17.70] | | 3 | 2.20 [0.60–17.40] | Kruskal-Wallis p = 0.0000 * --- CREATININE_min ---
c:\Users\home\anaconda3\envs\project1\Lib\site-packages\scipy\stats\_axis_nan_policy.py:531: UserWarning: scipy.stats.shapiro: For N > 5000, computed p-value may not be accurate. Current N is 1912223. res = hypotest_fun_out(*samples, **kwds)
| Class | Summary | |---------|------------------| | 0 | 1.40 [0.80–2.70] | | 1 | 1.80 [1.30–2.80] | | 2 | 1.20 [0.80–2.20] | | 3 | 0.80 [0.60–1.30] | Kruskal-Wallis p = 0.0000 * --- CREATININE_max ---
c:\Users\home\anaconda3\envs\project1\Lib\site-packages\scipy\stats\_axis_nan_policy.py:531: UserWarning: scipy.stats.shapiro: For N > 5000, computed p-value may not be accurate. Current N is 1912223. res = hypotest_fun_out(*samples, **kwds)
| Class | Summary | |---------|------------------| | 0 | 1.80 [1.10–3.20] | | 1 | 1.90 [1.50–3.10] | | 2 | 1.40 [1.00–2.60] | | 3 | 1.20 [0.80–1.80] | Kruskal-Wallis p = 0.0000 * --- CHLORIDE_min ---
c:\Users\home\anaconda3\envs\project1\Lib\site-packages\scipy\stats\_axis_nan_policy.py:531: UserWarning: scipy.stats.shapiro: For N > 5000, computed p-value may not be accurate. Current N is 1937638. res = hypotest_fun_out(*samples, **kwds)
| Class | Summary | |---------|------------------------| | 0 | 104.00 [98.00–109.00] | | 1 | 102.00 [102.00–109.00] | | 2 | 103.00 [98.00–107.00] | | 3 | 104.00 [99.00–107.00] | Kruskal-Wallis p = 0.0000 * --- CHLORIDE_max ---
c:\Users\home\anaconda3\envs\project1\Lib\site-packages\scipy\stats\_axis_nan_policy.py:531: UserWarning: scipy.stats.shapiro: For N > 5000, computed p-value may not be accurate. Current N is 1937638. res = hypotest_fun_out(*samples, **kwds)
| Class | Summary | |---------|------------------------| | 0 | 109.00 [104.00–114.00] | | 1 | 108.00 [105.00–111.00] | | 2 | 108.00 [102.00–112.00] | | 3 | 109.00 [105.00–113.00] | Kruskal-Wallis p = 0.0000 * --- GLUCOSE_min ---
c:\Users\home\anaconda3\envs\project1\Lib\site-packages\scipy\stats\_axis_nan_policy.py:531: UserWarning: scipy.stats.shapiro: For N > 5000, computed p-value may not be accurate. Current N is 1911814. res = hypotest_fun_out(*samples, **kwds)
| Class | Summary | |---------|-----------------------| | 0 | 100.00 [80.00–124.00] | | 1 | 75.00 [64.00–105.00] | | 2 | 101.00 [87.00–116.00] | | 3 | 106.00 [87.00–129.00] | Kruskal-Wallis p = 0.0000 * --- GLUCOSE_max ---
c:\Users\home\anaconda3\envs\project1\Lib\site-packages\scipy\stats\_axis_nan_policy.py:531: UserWarning: scipy.stats.shapiro: For N > 5000, computed p-value may not be accurate. Current N is 1911814. res = hypotest_fun_out(*samples, **kwds)
| Class | Summary | |---------|------------------------| | 0 | 156.00 [123.00–204.00] | | 1 | 114.00 [96.00–164.00] | | 2 | 154.00 [116.00–211.00] | | 3 | 183.00 [146.00–235.00] | Kruskal-Wallis p = 0.0000 * --- HEMATOCRIT_min ---
c:\Users\home\anaconda3\envs\project1\Lib\site-packages\scipy\stats\_axis_nan_policy.py:531: UserWarning: scipy.stats.shapiro: For N > 5000, computed p-value may not be accurate. Current N is 1968155. res = hypotest_fun_out(*samples, **kwds)
| Class | Summary | |---------|---------------------| | 0 | 26.20 [23.90–29.30] | | 1 | 23.30 [21.80–27.80] | | 2 | 24.80 [23.00–30.10] | | 3 | 27.00 [23.50–31.90] | Kruskal-Wallis p = 0.0000 * --- HEMATOCRIT_max ---
c:\Users\home\anaconda3\envs\project1\Lib\site-packages\scipy\stats\_axis_nan_policy.py:531: UserWarning: scipy.stats.shapiro: For N > 5000, computed p-value may not be accurate. Current N is 1968155. res = hypotest_fun_out(*samples, **kwds)
| Class | Summary | |---------|---------------------| | 0 | 31.60 [29.00–36.00] | | 1 | 28.80 [25.40–31.40] | | 2 | 32.70 [30.30–35.90] | | 3 | 35.30 [31.50–39.30] | Kruskal-Wallis p = 0.0000 * --- HEMOGLOBIN_min ---
c:\Users\home\anaconda3\envs\project1\Lib\site-packages\scipy\stats\_axis_nan_policy.py:531: UserWarning: scipy.stats.shapiro: For N > 5000, computed p-value may not be accurate. Current N is 1965848. res = hypotest_fun_out(*samples, **kwds)
| Class | Summary | |---------|-------------------| | 0 | 8.70 [8.00–9.80] | | 1 | 8.10 [7.50–9.60] | | 2 | 9.10 [7.90–10.40] | | 3 | 9.20 [7.90–10.60] | Kruskal-Wallis p = 0.0000 * --- HEMOGLOBIN_max ---
c:\Users\home\anaconda3\envs\project1\Lib\site-packages\scipy\stats\_axis_nan_policy.py:531: UserWarning: scipy.stats.shapiro: For N > 5000, computed p-value may not be accurate. Current N is 1965848. res = hypotest_fun_out(*samples, **kwds)
| Class | Summary | |---------|---------------------| | 0 | 10.40 [9.50–11.90] | | 1 | 9.50 [8.50–10.40] | | 2 | 10.80 [9.80–12.10] | | 3 | 12.20 [10.50–13.20] | Kruskal-Wallis p = 0.0000 * --- LACTATE_min ---
c:\Users\home\anaconda3\envs\project1\Lib\site-packages\scipy\stats\_axis_nan_policy.py:531: UserWarning: scipy.stats.shapiro: For N > 5000, computed p-value may not be accurate. Current N is 1249761. res = hypotest_fun_out(*samples, **kwds)
| Class | Summary | |---------|------------------| | 0 | 1.80 [1.20–3.10] | | 1 | 1.20 [0.90–2.30] | | 2 | 1.60 [1.10–3.10] | | 3 | 1.50 [1.10–2.30] | Kruskal-Wallis p = 0.0000 * --- LACTATE_max ---
c:\Users\home\anaconda3\envs\project1\Lib\site-packages\scipy\stats\_axis_nan_policy.py:531: UserWarning: scipy.stats.shapiro: For N > 5000, computed p-value may not be accurate. Current N is 1249761. res = hypotest_fun_out(*samples, **kwds)
| Class | Summary | |---------|------------------| | 0 | 3.50 [1.80–8.20] | | 1 | 1.80 [0.90–3.50] | | 2 | 3.20 [1.80–4.30] | | 3 | 2.30 [1.70–4.80] | Kruskal-Wallis p = 0.0000 * --- PLATELET_min ---
c:\Users\home\anaconda3\envs\project1\Lib\site-packages\scipy\stats\_axis_nan_policy.py:531: UserWarning: scipy.stats.shapiro: For N > 5000, computed p-value may not be accurate. Current N is 1964764. res = hypotest_fun_out(*samples, **kwds)
| Class | Summary | |---------|-----------------------| | 0 | 100.00 [51.00–195.00] | | 1 | 256.00 [71.00–477.00] | | 2 | 143.00 [84.00–206.00] | | 3 | 200.00 [78.00–314.00] | Kruskal-Wallis p = 0.0000 * --- PLATELET_max ---
c:\Users\home\anaconda3\envs\project1\Lib\site-packages\scipy\stats\_axis_nan_policy.py:531: UserWarning: scipy.stats.shapiro: For N > 5000, computed p-value may not be accurate. Current N is 1964764. res = hypotest_fun_out(*samples, **kwds)
| Class | Summary | |---------|------------------------| | 0 | 151.00 [84.00–257.00] | | 1 | 302.00 [110.00–517.00] | | 2 | 194.00 [119.00–254.00] | | 3 | 241.00 [128.00–362.00] | Kruskal-Wallis p = 0.0000 * --- POTASSIUM_min ---
c:\Users\home\anaconda3\envs\project1\Lib\site-packages\scipy\stats\_axis_nan_policy.py:531: UserWarning: scipy.stats.shapiro: For N > 5000, computed p-value may not be accurate. Current N is 1938534. res = hypotest_fun_out(*samples, **kwds)
| Class | Summary | |---------|------------------| | 0 | 3.80 [3.40–4.10] | | 1 | 4.00 [3.60–4.60] | | 2 | 3.70 [3.50–4.20] | | 3 | 3.60 [3.20–4.00] | Kruskal-Wallis p = 0.0000 * --- POTASSIUM_max ---
c:\Users\home\anaconda3\envs\project1\Lib\site-packages\scipy\stats\_axis_nan_policy.py:531: UserWarning: scipy.stats.shapiro: For N > 5000, computed p-value may not be accurate. Current N is 1938534. res = hypotest_fun_out(*samples, **kwds)
| Class | Summary | |---------|------------------| | 0 | 4.70 [4.20–5.30] | | 1 | 4.70 [4.10–4.90] | | 2 | 4.40 [4.00–4.90] | | 3 | 4.50 [4.10–5.00] | Kruskal-Wallis p = 0.0000 * --- PTT_min ---
c:\Users\home\anaconda3\envs\project1\Lib\site-packages\scipy\stats\_axis_nan_policy.py:531: UserWarning: scipy.stats.shapiro: For N > 5000, computed p-value may not be accurate. Current N is 1814197. res = hypotest_fun_out(*samples, **kwds)
| Class | Summary | |---------|---------------------| | 0 | 34.40 [29.20–44.30] | | 1 | 35.00 [30.20–47.50] | | 2 | 34.90 [29.30–40.60] | | 3 | 28.70 [25.60–33.30] | Kruskal-Wallis p = 0.0000 * --- PTT_max ---
c:\Users\home\anaconda3\envs\project1\Lib\site-packages\scipy\stats\_axis_nan_policy.py:531: UserWarning: scipy.stats.shapiro: For N > 5000, computed p-value may not be accurate. Current N is 1814197. res = hypotest_fun_out(*samples, **kwds)
| Class | Summary | |---------|---------------------| | 0 | 41.30 [31.40–61.40] | | 1 | 35.60 [33.70–47.90] | | 2 | 43.00 [35.00–56.10] | | 3 | 35.90 [29.30–50.60] | Kruskal-Wallis p = 0.0000 * --- INR_min ---
c:\Users\home\anaconda3\envs\project1\Lib\site-packages\scipy\stats\_axis_nan_policy.py:531: UserWarning: scipy.stats.shapiro: For N > 5000, computed p-value may not be accurate. Current N is 1815749. res = hypotest_fun_out(*samples, **kwds)
| Class | Summary | |---------|------------------| | 0 | 1.40 [1.30–1.70] | | 1 | 1.40 [1.30–1.50] | | 2 | 1.50 [1.20–1.90] | | 3 | 1.20 [1.10–1.40] | Kruskal-Wallis p = 0.0000 * --- INR_max ---
c:\Users\home\anaconda3\envs\project1\Lib\site-packages\scipy\stats\_axis_nan_policy.py:531: UserWarning: scipy.stats.shapiro: For N > 5000, computed p-value may not be accurate. Current N is 1815749. res = hypotest_fun_out(*samples, **kwds)
| Class | Summary | |---------|------------------| | 0 | 1.60 [1.30–2.20] | | 1 | 1.40 [1.30–1.60] | | 2 | 1.90 [1.40–2.20] | | 3 | 1.50 [1.20–1.90] | Kruskal-Wallis p = 0.0000 * --- PT_min ---
c:\Users\home\anaconda3\envs\project1\Lib\site-packages\scipy\stats\_axis_nan_policy.py:531: UserWarning: scipy.stats.shapiro: For N > 5000, computed p-value may not be accurate. Current N is 1815748. res = hypotest_fun_out(*samples, **kwds)
| Class | Summary | |---------|---------------------| | 0 | 15.70 [14.50–17.90] | | 1 | 15.40 [14.40–16.70] | | 2 | 14.90 [14.00–19.60] | | 3 | 13.70 [12.80–15.40] | Kruskal-Wallis p = 0.0000 * --- PT_max ---
c:\Users\home\anaconda3\envs\project1\Lib\site-packages\scipy\stats\_axis_nan_policy.py:531: UserWarning: scipy.stats.shapiro: For N > 5000, computed p-value may not be accurate. Current N is 1815748. res = hypotest_fun_out(*samples, **kwds)
| Class | Summary | |---------|---------------------| | 0 | 16.80 [15.10–21.70] | | 1 | 15.80 [14.60–18.00] | | 2 | 17.00 [15.00–22.10] | | 3 | 15.40 [13.90–19.40] | Kruskal-Wallis p = 0.0000 * --- SODIUM_min ---
c:\Users\home\anaconda3\envs\project1\Lib\site-packages\scipy\stats\_axis_nan_policy.py:531: UserWarning: scipy.stats.shapiro: For N > 5000, computed p-value may not be accurate. Current N is 1938003. res = hypotest_fun_out(*samples, **kwds)
| Class | Summary | |---------|------------------------| | 0 | 136.00 [132.00–140.00] | | 1 | 136.00 [132.00–139.00] | | 2 | 135.00 [131.00–137.00] | | 3 | 135.00 [131.00–138.00] | Kruskal-Wallis p = 0.0000 * --- SODIUM_max ---
c:\Users\home\anaconda3\envs\project1\Lib\site-packages\scipy\stats\_axis_nan_policy.py:531: UserWarning: scipy.stats.shapiro: For N > 5000, computed p-value may not be accurate. Current N is 1938003. res = hypotest_fun_out(*samples, **kwds)
| Class | Summary | |---------|------------------------| | 0 | 140.00 [136.00–143.00] | | 1 | 140.00 [135.00–143.00] | | 2 | 138.00 [133.00–141.00] | | 3 | 140.00 [137.00–143.00] | Kruskal-Wallis p = 0.0000 * --- BUN_min ---
c:\Users\home\anaconda3\envs\project1\Lib\site-packages\scipy\stats\_axis_nan_policy.py:531: UserWarning: scipy.stats.shapiro: For N > 5000, computed p-value may not be accurate. Current N is 1912022. res = hypotest_fun_out(*samples, **kwds)
| Class | Summary | |---------|---------------------| | 0 | 34.00 [18.00–58.00] | | 1 | 32.00 [23.00–66.00] | | 2 | 16.00 [12.00–31.00] | | 3 | 19.00 [15.00–37.00] | Kruskal-Wallis p = 0.0000 * --- BUN_max ---
c:\Users\home\anaconda3\envs\project1\Lib\site-packages\scipy\stats\_axis_nan_policy.py:531: UserWarning: scipy.stats.shapiro: For N > 5000, computed p-value may not be accurate. Current N is 1912022. res = hypotest_fun_out(*samples, **kwds)
| Class | Summary | |---------|---------------------| | 0 | 40.00 [23.00–66.00] | | 1 | 39.00 [28.00–71.00] | | 2 | 21.00 [16.00–35.00] | | 3 | 26.00 [19.00–43.00] | Kruskal-Wallis p = 0.0000 * --- WBC_min ---
c:\Users\home\anaconda3\envs\project1\Lib\site-packages\scipy\stats\_axis_nan_policy.py:531: UserWarning: scipy.stats.shapiro: For N > 5000, computed p-value may not be accurate. Current N is 1966221. res = hypotest_fun_out(*samples, **kwds)
| Class | Summary | |---------|--------------------| | 0 | 9.60 [4.20–13.90] | | 1 | 8.80 [5.70–19.00] | | 2 | 10.30 [6.40–16.20] | | 3 | 7.00 [5.10–12.50] | Kruskal-Wallis p = 0.0000 * --- WBC_max ---
c:\Users\home\anaconda3\envs\project1\Lib\site-packages\scipy\stats\_axis_nan_policy.py:531: UserWarning: scipy.stats.shapiro: For N > 5000, computed p-value may not be accurate. Current N is 1966221. res = hypotest_fun_out(*samples, **kwds)
| Class | Summary | |---------|--------------------| | 0 | 12.20 [6.20–19.90] | | 1 | 13.20 [8.80–21.70] | | 2 | 15.50 [9.60–19.00] | | 3 | 12.00 [7.70–16.70] | Kruskal-Wallis p = 0.0000 * --- SOFA ---
c:\Users\home\anaconda3\envs\project1\Lib\site-packages\scipy\stats\_axis_nan_policy.py:531: UserWarning: scipy.stats.shapiro: For N > 5000, computed p-value may not be accurate. Current N is 2276938. res = hypotest_fun_out(*samples, **kwds)
| Class | Summary | |---------|-------------------| | 0 | 9.00 [6.00–11.00] | | 1 | 7.00 [4.00–9.00] | | 2 | 6.00 [4.00–8.00] | | 3 | 5.00 [3.00–7.00] | Kruskal-Wallis p = 0.0000 * --- HeartRate_Min ---
c:\Users\home\anaconda3\envs\project1\Lib\site-packages\scipy\stats\_axis_nan_policy.py:531: UserWarning: scipy.stats.shapiro: For N > 5000, computed p-value may not be accurate. Current N is 1918917. res = hypotest_fun_out(*samples, **kwds)
| Class | Summary | |---------|----------------------| | 0 | 80.00 [65.00–92.00] | | 1 | 85.00 [80.00–100.00] | | 2 | 65.00 [58.00–84.00] | | 3 | 72.00 [52.00–96.00] | Kruskal-Wallis p = 0.0000 * --- HeartRate_Max ---
c:\Users\home\anaconda3\envs\project1\Lib\site-packages\scipy\stats\_axis_nan_policy.py:531: UserWarning: scipy.stats.shapiro: For N > 5000, computed p-value may not be accurate. Current N is 1918917. res = hypotest_fun_out(*samples, **kwds)
| Class | Summary | |---------|------------------------| | 0 | 119.00 [108.00–139.00] | | 1 | 126.00 [113.00–131.00] | | 2 | 99.00 [85.00–120.00] | | 3 | 111.00 [94.00–139.00] | Kruskal-Wallis p = 0.0000 * --- HeartRate_Mean ---
c:\Users\home\anaconda3\envs\project1\Lib\site-packages\scipy\stats\_axis_nan_policy.py:531: UserWarning: scipy.stats.shapiro: For N > 5000, computed p-value may not be accurate. Current N is 1918917. res = hypotest_fun_out(*samples, **kwds)
| Class | Summary | |---------|-----------------------| | 0 | 100.54 [87.13–116.91] | | 1 | 105.36 [93.68–112.88] | | 2 | 85.50 [67.96–95.91] | | 3 | 91.35 [71.31–116.25] | Kruskal-Wallis p = 0.0000 * --- SysBP_Min ---
c:\Users\home\anaconda3\envs\project1\Lib\site-packages\scipy\stats\_axis_nan_policy.py:531: UserWarning: scipy.stats.shapiro: For N > 5000, computed p-value may not be accurate. Current N is 1876734. res = hypotest_fun_out(*samples, **kwds)
| Class | Summary | |---------|----------------------| | 0 | 80.00 [70.00–94.00] | | 1 | 89.00 [81.00–107.00] | | 2 | 87.00 [80.00–102.00] | | 3 | 87.00 [81.00–95.00] | Kruskal-Wallis p = 0.0000 * --- SysBP_Max ---
c:\Users\home\anaconda3\envs\project1\Lib\site-packages\scipy\stats\_axis_nan_policy.py:531: UserWarning: scipy.stats.shapiro: For N > 5000, computed p-value may not be accurate. Current N is 1876734. res = hypotest_fun_out(*samples, **kwds)
| Class | Summary | |---------|------------------------| | 0 | 146.00 [126.00–161.00] | | 1 | 145.00 [130.00–150.00] | | 2 | 138.00 [126.00–154.00] | | 3 | 139.00 [128.00–160.00] | Kruskal-Wallis p = 0.0000 * --- SysBP_Mean ---
c:\Users\home\anaconda3\envs\project1\Lib\site-packages\scipy\stats\_axis_nan_policy.py:531: UserWarning: scipy.stats.shapiro: For N > 5000, computed p-value may not be accurate. Current N is 1876734. res = hypotest_fun_out(*samples, **kwds)
| Class | Summary | |---------|------------------------| | 0 | 108.91 [99.30–117.61] | | 1 | 110.78 [106.17–136.37] | | 2 | 111.81 [101.64–123.16] | | 3 | 110.93 [103.90–120.61] | Kruskal-Wallis p = 0.0000 * --- DiasBP_Min ---
c:\Users\home\anaconda3\envs\project1\Lib\site-packages\scipy\stats\_axis_nan_policy.py:531: UserWarning: scipy.stats.shapiro: For N > 5000, computed p-value may not be accurate. Current N is 1876729. res = hypotest_fun_out(*samples, **kwds)
| Class | Summary | |---------|---------------------| | 0 | 39.00 [30.00–46.00] | | 1 | 42.00 [30.00–46.00] | | 2 | 40.00 [32.00–48.00] | | 3 | 40.00 [36.00–49.00] | Kruskal-Wallis p = 0.0000 * --- DiasBP_Max ---
c:\Users\home\anaconda3\envs\project1\Lib\site-packages\scipy\stats\_axis_nan_policy.py:531: UserWarning: scipy.stats.shapiro: For N > 5000, computed p-value may not be accurate. Current N is 1876729. res = hypotest_fun_out(*samples, **kwds)
| Class | Summary | |---------|----------------------| | 0 | 87.00 [75.00–100.00] | | 1 | 82.00 [74.00–103.00] | | 2 | 77.00 [66.00–93.00] | | 3 | 84.00 [69.00–94.00] | Kruskal-Wallis p = 0.0000 * --- DiasBP_Mean ---
c:\Users\home\anaconda3\envs\project1\Lib\site-packages\scipy\stats\_axis_nan_policy.py:531: UserWarning: scipy.stats.shapiro: For N > 5000, computed p-value may not be accurate. Current N is 1876729. res = hypotest_fun_out(*samples, **kwds)
| Class | Summary | |---------|---------------------| | 0 | 58.42 [51.03–66.94] | | 1 | 56.48 [53.81–64.00] | | 2 | 55.69 [50.21–63.79] | | 3 | 59.65 [54.16–67.20] | Kruskal-Wallis p = 0.0000 * --- MeanBP_Min ---
c:\Users\home\anaconda3\envs\project1\Lib\site-packages\scipy\stats\_axis_nan_policy.py:531: UserWarning: scipy.stats.shapiro: For N > 5000, computed p-value may not be accurate. Current N is 1877314. res = hypotest_fun_out(*samples, **kwds)
| Class | Summary | |---------|---------------------| | 0 | 52.00 [39.00–60.00] | | 1 | 56.00 [48.00–61.00] | | 2 | 55.00 [47.00–62.67] | | 3 | 54.00 [50.00–61.00] | Kruskal-Wallis p = 0.0000 * --- MeanBP_Max ---
c:\Users\home\anaconda3\envs\project1\Lib\site-packages\scipy\stats\_axis_nan_policy.py:531: UserWarning: scipy.stats.shapiro: For N > 5000, computed p-value may not be accurate. Current N is 1877314. res = hypotest_fun_out(*samples, **kwds)
| Class | Summary | |---------|-----------------------| | 0 | 101.00 [89.00–114.00] | | 1 | 95.00 [88.00–115.00] | | 2 | 97.00 [85.00–106.67] | | 3 | 100.00 [88.00–113.00] | Kruskal-Wallis p = 0.0000 * --- MeanBP_Mean ---
c:\Users\home\anaconda3\envs\project1\Lib\site-packages\scipy\stats\_axis_nan_policy.py:531: UserWarning: scipy.stats.shapiro: For N > 5000, computed p-value may not be accurate. Current N is 1877314. res = hypotest_fun_out(*samples, **kwds)
| Class | Summary | |---------|---------------------| | 0 | 72.88 [66.96–79.74] | | 1 | 72.35 [67.69–80.04] | | 2 | 71.69 [66.00–79.54] | | 3 | 75.31 [71.21–81.29] | Kruskal-Wallis p = 0.0000 * --- RespRate_Min ---
c:\Users\home\anaconda3\envs\project1\Lib\site-packages\scipy\stats\_axis_nan_policy.py:531: UserWarning: scipy.stats.shapiro: For N > 5000, computed p-value may not be accurate. Current N is 1876702. res = hypotest_fun_out(*samples, **kwds)
| Class | Summary | |---------|---------------------| | 0 | 13.00 [10.00–16.00] | | 1 | 17.00 [10.00–20.00] | | 2 | 11.00 [9.00–14.00] | | 3 | 11.00 [10.00–14.00] | Kruskal-Wallis p = 0.0000 * --- RespRate_Max ---
c:\Users\home\anaconda3\envs\project1\Lib\site-packages\scipy\stats\_axis_nan_policy.py:531: UserWarning: scipy.stats.shapiro: For N > 5000, computed p-value may not be accurate. Current N is 1876702. res = hypotest_fun_out(*samples, **kwds)
| Class | Summary | |---------|---------------------| | 0 | 32.00 [26.00–36.00] | | 1 | 31.00 [26.00–36.00] | | 2 | 25.00 [22.00–30.00] | | 3 | 27.00 [22.00–31.00] | Kruskal-Wallis p = 0.0000 * --- RespRate_Mean ---
c:\Users\home\anaconda3\envs\project1\Lib\site-packages\scipy\stats\_axis_nan_policy.py:531: UserWarning: scipy.stats.shapiro: For N > 5000, computed p-value may not be accurate. Current N is 1876702. res = hypotest_fun_out(*samples, **kwds)
| Class | Summary | |---------|---------------------| | 0 | 21.68 [17.92–24.74] | | 1 | 24.84 [17.94–27.04] | | 2 | 17.10 [14.85–20.08] | | 3 | 17.97 [15.35–20.58] | Kruskal-Wallis p = 0.0000 * --- TempC_Min ---
c:\Users\home\anaconda3\envs\project1\Lib\site-packages\scipy\stats\_axis_nan_policy.py:531: UserWarning: scipy.stats.shapiro: For N > 5000, computed p-value may not be accurate. Current N is 1867072. res = hypotest_fun_out(*samples, **kwds)
| Class | Summary | |---------|---------------------| | 0 | 35.67 [35.22–36.39] | | 1 | 36.00 [35.56–36.67] | | 2 | 36.11 [35.60–36.56] | | 3 | 36.28 [35.72–36.67] | Kruskal-Wallis p = 0.0000 * --- TempC_Max ---
c:\Users\home\anaconda3\envs\project1\Lib\site-packages\scipy\stats\_axis_nan_policy.py:531: UserWarning: scipy.stats.shapiro: For N > 5000, computed p-value may not be accurate. Current N is 1867072. res = hypotest_fun_out(*samples, **kwds)
| Class | Summary | |---------|---------------------| | 0 | 37.33 [36.72–38.11] | | 1 | 38.28 [37.33–38.78] | | 2 | 37.17 [36.80–37.94] | | 3 | 37.44 [37.17–38.06] | Kruskal-Wallis p = 0.0000 * --- TempC_Mean ---
c:\Users\home\anaconda3\envs\project1\Lib\site-packages\scipy\stats\_axis_nan_policy.py:531: UserWarning: scipy.stats.shapiro: For N > 5000, computed p-value may not be accurate. Current N is 1867072. res = hypotest_fun_out(*samples, **kwds)
| Class | Summary | |---------|---------------------| | 0 | 36.65 [36.01–37.12] | | 1 | 37.09 [36.65–37.48] | | 2 | 36.70 [36.33–37.19] | | 3 | 36.85 [36.42–37.29] | Kruskal-Wallis p = 0.0000 * --- SpO2_Min ---
c:\Users\home\anaconda3\envs\project1\Lib\site-packages\scipy\stats\_axis_nan_policy.py:531: UserWarning: scipy.stats.shapiro: For N > 5000, computed p-value may not be accurate. Current N is 1876949. res = hypotest_fun_out(*samples, **kwds)
| Class | Summary | |---------|---------------------| | 0 | 92.00 [86.00–94.00] | | 1 | 90.00 [89.00–94.00] | | 2 | 95.00 [92.00–96.00] | | 3 | 94.00 [91.00–97.00] | Kruskal-Wallis p = 0.0000 * --- SpO2_Max ---
c:\Users\home\anaconda3\envs\project1\Lib\site-packages\scipy\stats\_axis_nan_policy.py:531: UserWarning: scipy.stats.shapiro: For N > 5000, computed p-value may not be accurate. Current N is 1876949. res = hypotest_fun_out(*samples, **kwds)
| Class | Summary | |---------|------------------------| | 0 | 100.00 [100.00–100.00] | | 1 | 100.00 [100.00–100.00] | | 2 | 100.00 [100.00–100.00] | | 3 | 100.00 [100.00–100.00] | Kruskal-Wallis p = 0.0000 * --- SpO2_Mean ---
c:\Users\home\anaconda3\envs\project1\Lib\site-packages\scipy\stats\_axis_nan_policy.py:531: UserWarning: scipy.stats.shapiro: For N > 5000, computed p-value may not be accurate. Current N is 1876949. res = hypotest_fun_out(*samples, **kwds)
| Class | Summary | |---------|---------------------| | 0 | 97.70 [96.10–98.82] | | 1 | 97.04 [96.60–98.33] | | 2 | 98.32 [97.35–99.26] | | 3 | 98.46 [96.92–99.58] | Kruskal-Wallis p = 0.0000 * --- Glucose_Min ---
c:\Users\home\anaconda3\envs\project1\Lib\site-packages\scipy\stats\_axis_nan_policy.py:531: UserWarning: scipy.stats.shapiro: For N > 5000, computed p-value may not be accurate. Current N is 1873003. res = hypotest_fun_out(*samples, **kwds)
| Class | Summary | |---------|-----------------------| | 0 | 99.00 [74.00–120.00] | | 1 | 73.00 [64.00–109.00] | | 2 | 111.00 [88.00–123.00] | | 3 | 106.00 [88.00–129.00] | Kruskal-Wallis p = 0.0000 * --- Glucose_Max ---
c:\Users\home\anaconda3\envs\project1\Lib\site-packages\scipy\stats\_axis_nan_policy.py:531: UserWarning: scipy.stats.shapiro: For N > 5000, computed p-value may not be accurate. Current N is 1873003. res = hypotest_fun_out(*samples, **kwds)
| Class | Summary | |---------|------------------------| | 0 | 165.00 [128.00–237.00] | | 1 | 154.00 [119.00–240.00] | | 2 | 148.00 [120.00–212.00] | | 3 | 188.00 [146.00–210.00] | Kruskal-Wallis p = 0.0000 * --- Glucose_Mean ---
c:\Users\home\anaconda3\envs\project1\Lib\site-packages\scipy\stats\_axis_nan_policy.py:531: UserWarning: scipy.stats.shapiro: For N > 5000, computed p-value may not be accurate. Current N is 1873003. res = hypotest_fun_out(*samples, **kwds)
| Class | Summary | |---------|------------------------| | 0 | 125.23 [106.62–169.00] | | 1 | 114.00 [99.20–162.00] | | 2 | 133.50 [106.00–165.00] | | 3 | 145.00 [119.95–166.00] | Kruskal-Wallis p = 0.0000 * --- weight_first ---
c:\Users\home\anaconda3\envs\project1\Lib\site-packages\scipy\stats\_axis_nan_policy.py:531: UserWarning: scipy.stats.shapiro: For N > 5000, computed p-value may not be accurate. Current N is 1922363. res = hypotest_fun_out(*samples, **kwds)
| Class | Summary | |---------|---------------------| | 0 | 80.20 [65.00–98.40] | | 1 | 74.00 [68.00–86.00] | | 2 | 84.20 [78.90–94.50] | | 3 | 70.00 [59.50–86.90] | Kruskal-Wallis p = 0.0000 * --- weight_min ---
c:\Users\home\anaconda3\envs\project1\Lib\site-packages\scipy\stats\_axis_nan_policy.py:531: UserWarning: scipy.stats.shapiro: For N > 5000, computed p-value may not be accurate. Current N is 1922363. res = hypotest_fun_out(*samples, **kwds)
| Class | Summary | |---------|---------------------| | 0 | 79.60 [63.80–95.10] | | 1 | 74.00 [65.00–84.00] | | 2 | 84.20 [71.30–92.00] | | 3 | 69.85 [58.20–83.70] | Kruskal-Wallis p = 0.0000 * --- weight_max ---
c:\Users\home\anaconda3\envs\project1\Lib\site-packages\scipy\stats\_axis_nan_policy.py:531: UserWarning: scipy.stats.shapiro: For N > 5000, computed p-value may not be accurate. Current N is 1922363. res = hypotest_fun_out(*samples, **kwds)
| Class | Summary | |---------|----------------------| | 0 | 85.73 [68.40–104.90] | | 1 | 79.20 [71.00–87.80] | | 2 | 89.90 [82.03–100.70] | | 3 | 75.10 [60.00–96.10] | Kruskal-Wallis p = 0.0000 * --- height_first ---
c:\Users\home\anaconda3\envs\project1\Lib\site-packages\scipy\stats\_axis_nan_policy.py:531: UserWarning: scipy.stats.shapiro: For N > 5000, computed p-value may not be accurate. Current N is 679195. res = hypotest_fun_out(*samples, **kwds)
| Class | Summary | |---------|------------------------| | 0 | 170.18 [162.56–175.26] | | 1 | 165.10 [154.94–170.18] | | 2 | 165.10 [157.48–180.34] | | 3 | 167.64 [157.48–180.34] | Kruskal-Wallis p = 0.0000 * --- height_min ---
c:\Users\home\anaconda3\envs\project1\Lib\site-packages\scipy\stats\_axis_nan_policy.py:531: UserWarning: scipy.stats.shapiro: For N > 5000, computed p-value may not be accurate. Current N is 679329. res = hypotest_fun_out(*samples, **kwds)
| Class | Summary | |---------|------------------------| | 0 | 170.18 [162.56–175.26] | | 1 | 165.10 [154.94–170.18] | | 2 | 165.10 [157.48–180.34] | | 3 | 167.64 [157.48–180.34] | Kruskal-Wallis p = 0.0000 * --- height_max ---
c:\Users\home\anaconda3\envs\project1\Lib\site-packages\scipy\stats\_axis_nan_policy.py:531: UserWarning: scipy.stats.shapiro: For N > 5000, computed p-value may not be accurate. Current N is 679329. res = hypotest_fun_out(*samples, **kwds)
| Class | Summary | |---------|------------------------| | 0 | 170.18 [162.56–175.26] | | 1 | 165.10 [154.94–170.18] | | 2 | 165.10 [157.48–182.88] | | 3 | 167.64 [157.48–180.34] | Kruskal-Wallis p = 0.0000 * --- bmi ---
c:\Users\home\anaconda3\envs\project1\Lib\site-packages\scipy\stats\_axis_nan_policy.py:531: UserWarning: scipy.stats.shapiro: For N > 5000, computed p-value may not be accurate. Current N is 679032. res = hypotest_fun_out(*samples, **kwds)
| Class | Summary | |---------|---------------------| | 0 | 27.35 [24.17–30.08] | | 1 | 26.39 [23.13–30.08] | | 2 | 31.70 [23.68–35.71] | | 3 | 26.43 [19.06–32.88] | Kruskal-Wallis p = 0.0000 * --- days_to_death ---
c:\Users\home\anaconda3\envs\project1\Lib\site-packages\scipy\stats\_axis_nan_policy.py:531: UserWarning: scipy.stats.shapiro: For N > 5000, computed p-value may not be accurate. Current N is 1484424. res = hypotest_fun_out(*samples, **kwds)
| Class | Summary | |---------|---------------------------| | 0 | 11.00 [3.00–20.00] | | 1 | 250.00 [73.00–291.00] | | 2 | 936.00 [640.00–1035.00] | | 3 | 2001.00 [1996.00–2020.00] | Kruskal-Wallis p = 0.0000 *
In [ ]:
print("\n📊 CATEGORICAL VARIABLES ANALYSIS\n")
from scipy.stats import chi2_contingency
for var in categorical_vars:
print(f"\n--- {var} ---")
contingency = pd.crosstab(analysis_df[var], analysis_df['outcome_class'])
chi2, p, _, _ = chi2_contingency(contingency)
# Show % distribution per group
perc_table = contingency.div(contingency.sum(axis=0), axis=1) * 100
print("Distribution (%):")
print(tabulate(perc_table.round(1), headers="keys", tablefmt="github"))
print(f"Chi-square p = {p:.4f} {'*' if p < 0.05 else ''}")